CN101878659A

CN101878659A - Hearing aid and a method of managing a logging device

Info

Publication number: CN101878659A
Application number: CN2007801017510A
Authority: CN
Inventors: J·奈尔森
Original assignee: Widex AS
Current assignee: Widex AS
Priority date: 2007-11-29
Filing date: 2007-11-29
Publication date: 2010-11-03
Anticipated expiration: 2027-11-29
Also published as: EP2215857A1; CA2706277A1; JP5031102B2; US20100232633A1; EP2215857B1; US8411888B2; DK2215857T3; CN101878659B; AU2007361787A1; CA2706277C; AU2007361787B2; JP2011504696A; WO2009068028A1

Abstract

A hearing aid (1) and a method of managing data stored in a logging device in a hearing aid is devised. The method involves obtaining sets of data representing parameters such as the sound environment at a predetermined data acquisition rate and storing the data according to one of a plurality of a set of possible parameter sets in a histogram (15) is having room for a limited number of instances of each particular set of parameters. If the limit for a particular histogram bin is reached when an instance is stored in the histogram, the data acquisition rate is adjusted according to a specified scheme and the number of instances of every parameter set is reduced by a fixed factor. Acquired data in the histogram thus reflects the most recent sound environments experienced by the hearing aid user.

Description

The method of hearing aids and management accounts equipment

Technical field

The present invention relates to hearing aids.More specifically, relate to digital deaf-aid, it comprises the device that is used for writing down the use recording parameters relevant with the hearing aids performance parameter with acoustic environment.

Background technology

Now, according to the regulation that is intended to eliminate for hearing impaired persons hearing loss, digital deaf-aid comprises and being used to handle with the precision of voice emplifying and complicated signal processing unit.In order to finely tune regulation setting (prescription setting), the statistical information of collecting about sound event from listen to environment is useful, and the specific hearing aids of expectation is listened in the environment at this and played a role in listening to environment.This information can preferably be stored in the hearing aids, therefore comprises that record (logging) equipment of non-volatile memory device is included in the hearing aids.Hereinafter, this is called as the hearing aids daily record.Parameter value was sampled with the daily record sampling interval, and set up the reflection of listening to environment that daily use of hearing aids and user in use run at leisure in the hearing aids daily record.

In this application, unless otherwise noted, term " daily record sampling (log sample) ", should be called as and measure and record (logging) parameter value, be enough to obtain on enough time spans of the main sound environment classification of some form at least, for example a few minutes magnitude the time interval, this measure and recording parameters value selected record in the hearing aids daily record.The daily record sampling period is also referred to as the acoustic environment sampling, and it is significantly greater than the input sample cycle, and the aanalogvoltage of expression sound pressure level is determined by this sampling period in input modulus (A/D) transducer.Be used for of the speed work of the existing input analog-to-digital converter of sound with for example 16-96kHz.The hearing aids type of discussing among the application is digital deaf-aid preferably, and wherein digital signal processor is carried out the adjusting and the amplification of sound for the user.This class hearing aids is divided into a plurality of separate bands with corresponding a plurality of band pass filters with signal usually.Each frequency band can be by independent amplification, and compression and noise reduce etc. and can carry out on each frequency band.

The hearing aids recording equipment illustrates in WO-A1-2007045276.These equipment master record two class incidents, that is, the user utilizes the time of specific program in the hearing aids, is called the statistical parameter record of service recorder and sign acoustic environment, is called histogram record (histogramlogging).

The histogram record is to come work by the event count in each histogram handle post (histogram bin) of accumulative total, as long as handle post (bin) is full, then increase intra-record slack byte by the selected factor, and reduce counting in all histogram handle posts with inverse factor (inverse factor), promptly effectively counter is reset (rebase) and write down this replacement.The mode of recording voice incident produces the histogram in the record cycle of expression prolongation like this.

Record data can include but not limited to characterize the data of listening to environment, about the data of user's hearing aids operation, be the variation that volume is provided with, and the variation in the hearing aids between distinct program and about the data of hearing aids built-in function.The combination of different event type also can be considered in record, and the user switches to specific program in the situation as listening at some.

The hearing aids recording equipment comprise according to predefine represent acoustic environment might parameter combinations histogram, each parameter combinations is represented by specific handle post in the histogram.Acoustic environment is with specific interval sampling, and nearest corresponding handle post is tired increases the appearance of record specific sound environment in the hearing aids daily record.

The content of daily record mainly is used in the assembled condition, wherein the hearing aids assembly crewman extracts data from the memory of hearing aids recording equipment, and the meeting hearing aid user, grasp about the user and listen to the experience of using the hearing aids of Set For Current during writing down in the situation specific.When daily record data relatively and user recall listen to situation the time, the memory of hearing aid user may make the user can not recall the specific situation of listening to of short duration, and for example may write down before several weeks and so incident of listening to of being forgotten for a long time by the user.This can make the assembly crewman produce some to obscure, and can cause the assembly crewman unnecessarily to change the setting of hearing aids.As a result, hearing aids may poorly be optimized, and adjustment is exactly the time of wasting the assembly crewman, and therefore causes user's discomfort.

Recording voice self in hearing aids not, this is because it requires in the hearing aids the restricted hardly so that stored sound of memory span, therefore several specific characters of stored sound only.The characteristics that the decision of two main standard will be stored just are provided with the metrizability and the level of relevant intrinsic information with hearing aids.

Experience shows and comprises that being recorded in of three parameters reaches suitable balance between storage economy and the level of detail, these three parameters be the noise level of expression sound first parameter, expression sound modulation level second parameter and the 3rd parameter of noise spectrum slope in the sound is described.

Noise level is defined as background noise level, and is by measuring at the envelope of sound event average 10% percentage on the sampling period.This noise level provides valuable information about current noise average level in the signal for signal processor in the hearing aids, and noise level also can be the assembly crewman provides the noise level that experiences about the user in using the hearing aids process information.

The useful signal amount that modulation level is defined as changing, and the envelope level by measuring 90% percentage and determine from the envelope level that the envelope level of sound event 90% average percentage on the sampling period, deducts 10% percentage of measurement.Modulation level mainly is used for determining the existence of voice in the signal by the hearing aid signal processor, and it also can be the useful information that the assembly crewman provides the acoustic environment character that experiences about hearing aid user.

Can be by the envelope level of 10% percentage of each frequency band in average a plurality of frequency bands, and determine the slope of resulting linear average to frequency axis, come the slope of calculating noise spectrum.This slope be each input sample and sound event on the sampling period averaged result calculate once.The slope of noise spectrum allow the hearing aid signal processor with the qualitative classification of noise so that optimize the computing of noise reduction algorithms in the hearing aids, listen artifact and maximum noise to reduce so that realize I, and the assembly crewman can obtain Useful Information from this noise spectrum slope knowledge, appears in the acoustic environment that is experienced so that determine whether the noise of some type.

In the use, three parameters are by continuous measurement, and the average level of measured value is stored in the buffer.In the sound event sampling period, thereby buffer contents is analyzed sampling is categorized as a plurality of possible acoustic environments, and corresponding histogram handle post record adds up in the hearing aids record, buffer resets then, in this mode and on certain hour, the histogram of the different possibility of expression acoustic environment frequency is set up in the hearing aids recording equipment.

In the section, three parameter collection are in the vector of the average acoustic environment of expression at the fixed time.The vector of expression acoustic environment is as the record storage, so that subsequent analysis.A plurality of possible acoustic environments can be detected by system, and are preset as a plurality of initial empty handle post in the memory of distribution, and the set of handle post forms histogram.

Daily record can contain a specific once appearance of listening to incident, 15 appearance of another incident and the incident of more frequent appearance.If aiming at having write down in several time-of-weeks a back incident hearing aids day occurs for 42 times, but the space of only having distributed 15 countings, then the counter about back one incident should will reach limit value, therefore and the balance multilated of different event in the daily record should be provided with higher weight to the individual event about the incident of more frequent appearance.The following describes the daily record overflow problem.

According to prior art, the daily record overflow problem solves by following steps, and promptly a handle post reaches may counting of maximum number in histogram, for example occurs for 15 of the specific sound incident times, just to histogram down-sampled (decimate).This be by with the content of all handle posts in the histogram divided by 2 and for sample rate is reduced by half so that subsequent sampling normalization record data are realized.

Yet the mode of management hearing aids daily record has at least two problems of not expecting.First problem is that the specific sound environment that record repeatedly and not writes down in the aft section in record cycle in the initial part in record cycle is stored in the histogram, and those in fact uninterested acoustic environments are provided with important weight.Second problem is that daily record has applied strict time restriction to hearing aids, this is because reach minimum possible sample rate after continuously down-sampled, perhaps because the data of record are more and more inaccurate and unreliable, this is owing to cause as several skewed popularity record cases in conjunction with first problem definition.

Summary of the invention

Therefore need a kind of method of managing the hearing aids daily record as follows, that is, emphasize to support the new data of historical data, and allow recording voice environment on the uncertain time period.

Further aim of the present invention is a kind of hearing aids of design, and it can be random time and the mode record data of experiencing with better hearing aid user.

Non-volatile memory block is to limit according to the write operation number of times that is allowed.Further aim of the present invention is a kind of hearing aids of design, and it can handle detailed daily record and can store data in nonvolatile memory on the section in the service time that prolongs.

In first aspect, hearing aids according to the present invention comprises the input transducer that produces input signal, thereby handle the hearing aids processor that input signal produces output signal, respond the output transducer of described output signal and have analyzer, the recording equipment of timer and memory, this memory has one group of histogram counters about predetermined group acoustic environment, wherein said analyzer is handled input signal and classify sound event between predetermined group acoustic environment, wherein timer triggers the output of the classification of sound event, wherein said memory receives classification and adds up about the counter of acoustic environment, wherein memory has the detection that overflow detector is used to monitor histogram counters and responds the incident of overflowing, reset except that described content and to all counters with predetermined factor, and wherein memory has the timer decision logic and is used for control timer, and this timer decision logic is by determining that timer setting responds resetting event.

Represent garbage owing to recognize the actual number that incident occurs in the hearing aids daily record at any given time, and the relative magnitude (magnitude) between the incident of different recording is a Useful Information more, therefore realizes that the suitable method of this knowledge is the average principle of exponential in the management of hearing aids recording equipment.

In second aspect, the invention provides method according to claim 7.

Further feature and advantage can be by finding out in the dependent claims.

Description of drawings

Further describe the present invention below with reference to the accompanying drawings in detail, wherein

Fig. 1 is the schematic diagram with hearing aids of recording equipment according to the present invention;

Fig. 2 is the histogram example from the record data of hearing aids shown in Fig. 1;

Fig. 3 is the histogram example to record data among handle post counting replacement back Fig. 2; And

Fig. 4 is a flow chart of carrying out the inventive method algorithm.

Embodiment

Fig. 1 illustrates the block diagram of the hearing aids 1 with recording equipment 4 according to the present invention.Hearing aids 1 comprises input microphone 2, bank of filters 3, recording equipment 4, hearing aids processor 20, ∑-Δ (sigma-delta) modulator 21, output stage 22 and acoustics output transducer 23.Recording equipment 4 comprises input/

output interface piece

5,10% percentage piece (percentile block), 6,90% percentage piece 7, noise spectrum slope indicating device piece 8, middle sum block 9, daily record data preparation piece 10, timer piece 11, log store piece 12.Log store piece 12 comprises volatile memory piece 13 and non-volatile memory block 14.Non-volatile memory block 14 can be stored at least one Nogata Figure 15.Non-volatile memory block 14 has the output that is connected to analyzer piece 17 inputs.The output of analyzer piece 17 is connected to the input of sample rate controll block 16.The output of sample rate controll block 16 is connected to the control input end of timer piece 11.

In the process of assembling hearing aids 1, recording equipment 4 can start through input/output interface 5.Acoustic signal is picked up and is converted into the signal of telecommunication by hearing aid microphone 2.The output signal of microphone 2 is divided into two branches.A branch is fed into bank of filters 3 so that further handle another branch's feed-in recording equipment 4.The output signal of bank of filters 3 is fed into the input of hearing aids processor 20.Hearing aids processor 20 is carried out acoustic processing according to the regulation of eliminating auditorily handicapped, and the output of hearing aids processor 20 by feed-in sigma-delta modulator 21 and output stage 22 so that drive acoustics output transducer 23.

In recording equipment 4, input signal is divided into three branches in order to analyze.The overall noise level that comprises first branch decision entering signal of 10% percentage piece 6.Second branch that comprises 90% percentage piece 7 is used in combination with

middle summing junction

9 and 10% percentage piece 6, thereby determines the modulation of audio signal by calculating difference between 90% percentage and 10% percentage.Comprising that the 3rd branch of noise spectrum slope indicating device piece 8 is used for the slope of definite noise spectrum, that is, is that high frequency is dominant or low frequency is dominant in the noise.

Consider that in the lump the parameter sets that comprises noise level parameter, modulation level parameter and be expressed as the noise spectrum Slope Parameters of α is regarded as representing the suitable feature of acoustic enviroment at given time, and without actual storage sound self.After analysis, parameter set is provided for daily record data preparation piece 10, this daily record data preparation piece 10 is carried out normalization, quantize and will gather in the classification of three parametric classifications acoustic environment in a plurality of possible acoustic environments, wherein acoustic environment is represented by multidimensional vector, so that be stored among Nogata Figure 15.Timer piece 11 is used for determining the daily record sampling period, that is, the acoustic environment that 10 outputs of data preparation piece are determined is to the frequency of log store piece 12.

Daily record data preparation piece 10 provides definite acoustic environment to arrive the volatile memory piece 13 of log store piece 12.The acoustic environment of determining of blotter is wanted in 13 storages of volatile memory piece, and it can store the complete histogram 15 of the acoustic environment of all records, thereby make it possible to read, be used for detecting so that the content of daily record can be retrieved by input/output interface 5.The acoustic environment incident that comprises the record of predetermined number whenever volatile memory piece 13, volatile memory piece are just with in the content write non-volatile memory piece 14 among Nogata Figure 15.Because can be restricted the useful life of non-volatile memory block 14 according to the number of times of possible write operation, therefore preferred this method is so that prolong the Acceptable life of the element of hearing aids recording equipment 4.

The handle post (bin) of analyzer 17 in each histogram carried out the content analysis of Nogata Figure 15 when overflowing, and uses the information that is obtained to control sample rate controll block 16.According to the content of Nogata Figure 15, analyzer 17 provides information about the optional sampling rate so that the recording voice environmental data for sample rate controll block 16.When writing down at first through input/output interface 5 startups, be used for triggering the pulse rate of daily record datas preparation pieces 10 by timer piece 11, promptly sample rate is set at flank speed.Analyzer 17 can determine to reduce sample rate subsequently, and for example the handle post incident of overflowing by Nogata Figure 15 starts.

Consider such situation, wherein Nogata Figure 15 can deposit the appearance up to 16 specific sound environment, the sample rate operation that for example record can 1/16Hz, or in other words, the parameter of per 16 seconds acoustic environments of record in daily record.If equivalent environment of per 16 seconds records, then corresponding handle post will fill up after 16 daily record incidents just, and histogram will equal to produce after 4 minutes 16 seconds the incident of overflowing therefore at 256 seconds.After sending the incident of overflowing, histogram will be reset (rebase) and sample rate will be reduced, and preferably is reduced to half of initial sample rate, and therefore record will carry out with the speed of 1/32Hz, or per 32 seconds once, the new situation of recording voice environment event in the histogram after replacement.

I/O piece 5 is used for starting or stoping recording process, and is used for reading from Nogata Figure 15 of hearing aids recording equipment 4 data of storage.In normal use, after starting hearing aids recording equipment 4, I/O piece 5 does not turn round (inactive), as long as hearing aids 1 is opened and in use, then hearing aids recording equipment 4 continues the recording voice environment event at regular intervals.

Fig. 2 is the histogrammic graphical display of hearing aids daily record according to example previously discussed.Daily record comprises three parameters of change resolution, shown in the medium and small table of Fig. 2.Parametric representation can be from three different types of data of hearing aids input signal acquisition, two different noise slope value α, three different noise level value and three different modulating values.

Because each combination of parameter value all is unique, so daily record must solve the different parameter combinations of 2*3*3=18 kind, the appearance that different parameters makes up can be recorded in as shown in Figure 2 the histogram.Here, the handle post on the abscissa is with x, y, and the form mark of z, wherein x represents the noise slope, y represents noise level, and z represents modulation.What histogram reflection parameter was different may be combined in the frequency that occurs in frame preset time.In this example, the resolution of three parameters significantly lowers, so that simplify diagrammatic sketch.The parameter of physical record can have much higher resolution, and for example each parameter has 256 kinds of different values.

As seeing among Fig. 2, the generation of acoustic environment situation can be changed into another parameter combinations from a parameter combinations.For example,

combination

1,2,3 and 1,3,2 does not appear in the histogram, occurs 10 times in the frame in preset time and make up 1,2,1.Therefore histogram writes down the appearance of each possibility parameter combinations, and therefore writes down the result.In this example, the memory space of distributing to the hearing aids daily record can be stored each possibility parameter combinations appearance up to 16 times.In actual histogram, the appearance number of each possibility parameter combinations can increase arbitrarily.

When the daily record incident of overflowing takes place, that is, and when a histogram handle post overflows, by with the number of the record of storing in each handle post divided by common factor, for example 2 or 4, come whole histogram is reset, and the new number that writes down in each handle post in the histogram is stored in the histogram.Can not be rounded up a plurality of zero in the corresponding handle post in the histogram that single counting in the

handle post

1,1,3 can be mapped to the postpone of attaching most importance to of therefore resetting by the number that common factor is divided exactly.

In this example,

combination

2,1,3 causes that most possibly daily record is overflowed next time, because this is the combination of frequent record in the histogram.If just write down more twice appearance of this specific parameters combinations, then counter will overflow, and will reset and the reducing of subsequently sample rate.

A consideration about this recording voice environmental manner is, if sample rate is repeated to be reduced to certain below the point, then Ji Lu acoustic environment can be than long interval record, because the changing features of acoustic environment must be more true faster than the speed that daily record can be write down, and makes that between this interval the acoustic environment of record appears at arbitrarily in the resulting daily record histogram.Therefore acoustic environment with duration shorter than recording period can escape from the record that detects with subsequently, even it has some importance to hearing aid user.

Another consideration is that data older in the histogram still have same weight, though it is recorded before several weeks.If it is poor that daily record and user remember relevance, user memory has the trend of decline in time, as assembly crewman during from the daily record of hearing aids memory fetch, is difficult to explain data in the histogram by meaningful ways.

When writing down enough data in the hearing aids daily record, after several hours, second logging mode is activated usually.The replacement that also starts the hearing aids daily record is overflowed in daily record in second logging mode, but sample rate remains on fixed value.This can have such effect, and the importance of older data was reduced when promptly each daily record was overflowed, and therefore made the new data that is recorded in the hearing aids daily record relatively more preponderate.

Diagrammatic sketch according to daily record overflow problem solution of the present invention is shown in Figure 3, and the value (solid line illustrates) after all initial values (shown by dashed lines) of each the handle post in wherein similar to histogram among Fig. 2 histogram are reset after the handle post overflows replaces.Histogram is reset and is comprised all handle post values that reduce by half, though also can use other replacement schemes, as using handle post value divided by the factor 3 or 4.All even number handle post values are directly reduced by half, and all odd number handle post values are rounded to nearest even number value, reduce by half then.Therefore after resetting, histogram keeps handle post ratio relative to each other.

The sample rate that reduces by half after event histogram is reset, the step of this method is consistent with the step of art methods.Yet after event histogram was reset, value before sample rate remains on still kept handle post ratio relative to each other, but compared with the data that collect the back of resetting, and the relative weighting of the data of collecting before resetting will reduce.After the operation of resetting continuously, according to the parameter combinations that system detects, handle post ratio relative to each other reflects recent history in more progressive mode.Continuously resetting event will further reduce the weight of legacy data, so that its weight decays in time.

At first, if ignore the approximation error of introducing by resetting, the information of collecting from the histogram of resetting is identical with information available before resetting.The relative size that writes down in each handle post in the histogram is basic identical, and

parameter combinations

2,1,3 still the most normal appearance, and before resetting, the appearance number of other parameter combinations and

parameter combinations

2,1,3 have identical relation.

In some cases, the big variation in the acoustic environment of record can cause the inaccurate of daily record data.For example, if the user experiences many alternative sounds environment in recording period, then many handle posts can speed much at one be filled in the daily record.Yet if the user only experiences several alternative sounds environment in identical recording period, only one or two handle post is filled, and other handle posts remain sky.In first situation, a large amount of samplings of alternative sounds environment will occur, and therefore before a handle post overflows, consume long recording period.In second situation, suppose that sample rate is identical in two kinds of situations, then the handle post will overflow sooner than first situation.

For the image that obtains to change in the acoustic environment of user's experience in recording period, the method according to this invention is the near more big more importance of incident setting that writes down in the daily record.The weight of institute's recording events can be managed by the change histogram and be carried out, and its mode will be explained in more detail below.

As long as parameter combinations handle post fills up in the histogram, and therefore histogram wait for aforementioned replacement, will carry out two operation bidirectionals.First operation is whether the scanning histogram has the handle post of filling more than 3/4ths.In digital system, the most significant bit of handle post counting that can be by detecting each handle post and detect time significant bits then, and easily realize.If two positions all are set, then this specific handle post is to be filled into more than 3/4ths, and shows the identity of this handle post.Second operation is to be independent of histogram self to store this information, therefore requires to filling the handle post memory allocated space more than 3/4ths.

When storage was filled information more than 3/4ths about which handle post, hereinafter referred to as background information, histogrammic Distribution Statistics analysis can be based on this information and executing.This analyzes to produce about acoustic environment to change information how soon, and is used for determining to collect the sample rate of acoustic environment data.

Narrow distribution represents that one or several acoustic environment type is dominant in histogram, and acoustic environment is to spare matter (homogenous) relatively in time.The memory write incoming event can be preserved by reducing sample rate.It is relative nonuniformity (heterogenous) in time that wide distribution means acoustic environment.Therefore by increasing sample rate, the more accurate effect of the acoustic environment that can obtain to experience.After analyze adjusting sample rate based on this, histogram can down-sampled like that as described above (decimate).

When analyzing the sense data of hearing aids daily record, the hearing aids assembly crewman can collect Useful Information from the background information of histogram and storage.Histogram can provide the information about acoustic environment, as the level and the feature of background noise level, and the existence of the voice signal of signal percentage as a whole.The background information of storage can provide the information of the alternative sounds environmental change of experiencing about the user in whole recording period.

According to the initial expected of assembly crewman about acoustic environment, the acoustic environment of hearing aid user experience normally writes down on the time span of several thoughtful some months.Because when record occurred over just hearing aids and opens, operating time information was by (on-time) counter records of operating time in the hearing aids.This operating hours counter is used in combination with record data, so that set up the image of the acoustic environment that hearing aid user experiences in recording period.

The actual audio that recording process and hearing aids are carried out in the hearing aids is handled and is carried out simultaneously.In a preferred embodiment, hearing aids log record noise level, the slope of modulation level and noise spectrum, and about how operating the information of hearing aids, what program to be preferred as is set in any level with volume control, or the like, but but also record of other parameters.For example, activity and performance that the sound that exceeds comfort level (comfort level) upper limit occurs surpassing two seconds, feeds back cancellation system, pick-up coil or directly audio frequency input, or the like.Because the memory free memory of hearing aids is limited, can in the hearing aids daily record, store the data minimizing of carrying out some form before the data.

The sample rate preferred tunable of hearing aids executive logging is whole.Experiment shows that the sample rate between to 15 minute is satisfied when considering the record data details of balance desired level about storage economy.Sample rate can start record and by initial setting, also can be adjusted automatically by the simple analysis of carrying out histogram data and background information by the hearing aids processor by the hearing aids assembly crewman.

The operation of resetting in the event histogram is just pending, and background information shows histogram data wide dispersion (large spread), then can run into many alternative sounds environment.Then, sample rate is advantageously increased, thereby guarantees that the specific sound environment was recorded before it changes feature, and this is because acoustic environment may change in the sampling time section.

On the other hand, the operation of resetting in the event histogram is just pending, and background information shows that histogram data scatters (small spread) among a small circle, then can run into only several alternative sounds environment.Sample rate is advantageously reduced then, so that save memory, this is because acoustic environment can not change in the sampling time section.

Therefore the hearing aids daily record provides the quantitative information about the qualitative condition of work of hearing aids as writing down in the special time period for the hearing aids assembly crewman.This information can be used when speaking face to face with hearing aid user together, so that clarification is about adjusting the possible problem of hearing aids regulation.By understand the acoustic environment that is dominant that hearing aid user experiences in wearing and use the hearing aids process, the hearing aids assembly crewman can design better hearing aids assembling.

For example, if some listens to voice under the condition hearing aid user complaint indigestion, but be difficult to describe the concrete situation of the when and where that encounters difficulties, then the hearing aids assembly crewman can extract and analyze the hearing aids daily record, wear and use the acoustic environment that experiences in the hearing aids process so that determine the user, and can correspondingly take action based on the information that obtains from the hearing aids daily record and hearing aids assembly crewman self experience and adjust the hearing aids assembling.

In generic instance, hearing aid user may be complained the voice under some type noise of indigestion, but the user can not describe noise characteristic, can not remember to experience the accurate situation of this difficulty, may be owing to lacking suitable hearing vocabulary or forgetting.

The hearing aids assembly crewman start the record of acoustic environment, and hearing aid user returns to its normal daily routines by using special-purpose order startup hearing aids daily record in the hearing aids mechanical assembly software then.When hearing aid user returned after several weeks, for example the hearing aids daily record may disclose a great deal of high-frequency noise or fizz the situation that is dominant.Then the assembly crewman can utilize the knowledge about the exact nature that is stored in the acoustic environment that is experienced in the daily record, and can, for example adjust the hearing aids accessory, so that voice are dominant, thereby eliminate the difficulty that hearing aid user is understood voice in the specific sound environment of specific hearing aid user experience on higher frequency by other adjustable parameters in adjust frequency response, compressor reducer setting and the hearing aids.

The outward appearance that specific at any time histogram is read depends on sample rate.The additive method of control sample rate relates to the statistical analysis of histogram content, and this is more accurate than the content of only counting each handle post.The control sample rate is important, and its reason is explained below in more detail.

If hearing aid user experiences a large amount of alternative sounds environment in recording period, the histogram that then obtains has quite wide Distribution Statistics, and this is owing to it seems that many handle posts are equal fillings.Such situation can be discerned by histogram is used suitable statistical analysis.In this case, thus increasing sample rate obtains that more samplings of acoustic environment are useful in the similar recording period.In this mode, the more detailed image of the acoustic environment type of the actual experience of user can be found out from the histogram that obtains.

Yet if hearing aid user only experiences several alternative sounds environment in recording period, the histogram that obtains has the Distribution Statistics that is rather narrow, and this is owing to only several handle posts when carrying out the histogram replacement are full.Such situation also can be discerned by histogram is used suitable statistical analysis.In this case, thus reduce sample rate in similar recording period, obtain acoustic environment still less the sampling be useful.In this mode, the image of the less details of the acoustic environment type of the actual experience of user can be found out from the histogram that obtains.

The algorithm flow chart of the method for explanation management data collection and storage according to the present invention is shown in Figure 4.The purpose of this algorithm is data acquisition rate was reset and adjusted to explanation histogram handle post when full to histogram a situation, and data acquisition rate correspondingly is expressed as sample rate here.

This algorithm can be considered and is divided into two parts.First comprises step 101,102,103,104 and 105, it handles the data acquisition of acoustic environment incident, and second portion comprises step 106,107,108,109,110,111,112 and 113, and it is handled histogram analysis, sample rate adjustment and histogram handle post and resets.These tasks will be explained below in more detail.

Algorithm is set variable herein and also is histogram memory allocated space from step 101.In step 102, check whether input has new acoustic environment sampling.If not new sampling occurs, then enter etc. to be recycled by jumping to step 103.As long as the sampling of new acoustic environment is ready, then by jumping to step 104, with sample record in histogram.In step 104, after the record sampling, in step 105, carry out test, so that whether the histogram handle post of store sample fills up in the determining step 104.If discontented, then record continues, and algorithm is circulated back to step 102 so that wait for next sampling.

Yet if the histogram handle post of store sample fills up in step 104, algorithm jumps to the second portion of algorithm through step 106, carries out histogrammic statistical analysis here.One of analysis result is the histogram distribution analysis, promptly checks histogram so that determine whether to exist in these conditions one.

The first condition of being checked is so-called " narrow distribution " situation of checking in step 107.Narrow distribution shows that only several handle posts reach its maximum when handle post in the histogram is full in the histogram.This shows that only several acoustic environments are dominant in daily record.In other words, the acoustic environment of experience is relatively stable in time.In this case, sample rate can advantageously reduce, because the many acoustic environment incidents that are recorded in the histogram are basic identical.

If narrow distribution do not occur, then algorithm jumps to step 110 rapidly.If there is narrow distribution, then whether be minimum possible sample rate to algorithm if jumping to step 108 so that detect current sample rate.If not, then algorithm jumps to step 109, and wherein sample rate reduces, and algorithm is circulated back to step 113, and wherein all handle posts are reset, and as previously described, and enter step 102, so that wait for next sampling.Yet if sample rate is minimum possibility sample rate, algorithm is circulated back to step 113, and wherein the handle post is reset as previously described, and enters step 102 so that wait for next sampling.

The second condition of being checked is so-called " wide distribution " situation that detects in step 110.Wide distribution shows that many handle posts are near its maximum when handle post in the histogram is full in the histogram.This shows that many alternative sounds environment have been deposited with in the daily record, and in other words, the sound that is experienced has changed a lot in time.In this case, sample rate can advantageously increase, because many alternative sounds environment events are recorded in the histogram.

If wide distribution do not occur in the histogram of analyzing, then algorithm is from step 110 redirect, and circulation turns back to step 113, and all handle posts are reset as previously described herein, and enter step 102, so that wait for next sampling.

Whether if there is wide distribution, then algorithm jumps to step 111, be the highest possibility sample rate so that detect current sample rate.If not, then algorithm jumps to step 112, and sample rate increases herein, and the algorithm circulation turns back to step 113 and enter step 102, so that wait for next sampling.

Yet if sample rate has been in its minimum value, the algorithm circulation turns back to step 113 and enters step 102, so that wait for next sampling.In fact, the 3rd condition just, promptly histogram distribution is uncertain, and sample rate remains unchanged.

As long as reading by the hearing aids assembly crewman of hearing aids daily record carried out, then the relative appearance of possibility parameter combinations is true for the acoustic environment maintenance of hearing aid user actual experience in recording period in the hearing aids daily record, even the number of times that one or more parameter combinations has occurred is above the actual number of times that contains of daily record.Therefore the hearing aids daily record provide strong tool for the hearing aids assembly crewman in case the fine setting hearing aid user can with the program of listening to.

Claims

1. hearing aids, it comprises:

Produce the input transducer of input signal; Thereby handle the hearing aids processor that described input signal produces output signal; Respond the output transducer of described output signal; And have analyzer, the recording equipment of timer and memory, described memory has one group of histogram counters about predetermined group acoustic environment, wherein said analyzer is handled described input signal and the described sound event of classification between described predetermined group of acoustic environment, wherein said timer triggers the output of the classification of described sound event, wherein said memory receives described classification and the counter about described acoustic environment is added up, wherein said memory has overflow detector, described overflow detector monitors described histogram counters and responds the detection of the incident of overflowing, by divided by predetermined factor all counters being reset with content, and wherein said memory has the timer decision logic of the described timer of control, and described timer decision logic responds resetting event by determining described timer setting.

2. hearing aids according to claim 1 is characterized in that described overflow detector has the device that is used to analyze described histogram and sets the sample rate that reduces under narrow histogram distribution situation.

3. hearing aids according to claim 1 is characterized in that described overflow detector has the device that is used to analyze described histogram and sets the sample rate that increases under wide histogram distribution situation.

4. hearing aids according to claim 1, it is characterized in that described memory comprises volatile memory piece and non-volatile memory block, described memory is used for storing data at described volatile memory piece, and discontinuously will be from the data record of described volatile memory piece in described non-volatile memory block.

5. according to the described hearing aids of aforementioned each claim, it is characterized in that being recorded in the characteristic parameter that data sampling in the described daily record is represented described acoustic environment.

6. according to the described hearing aids of aforementioned each claim, it is characterized in that being used for analytic record comprises at the device of the described data of described daily record and is used for determining a plurality ofly may writing down the device that record of distribution distributes and being used for distributing based on determined record and producing a plurality of devices that may control signals.

7. method of managing data record in the hearing aids, said method comprising the steps of: with selected speed acquisition parameter data, with the data setting of described collection be stored in the histogram in the memory of distribution of described hearing aids, whether the appearance of the data that test is stored surpasses the pre-determined maximum number restriction, and in this case, reduce the number that all data occur pro rata with predetermined factor, analyze described histogram data, determine new data acquisition sample rate based on described analysis, and correspondingly change selected sample rate.

8. method according to claim 7, it is characterized in that described collection supplemental characteristic at least some the expression acoustic environments.

9. method according to claim 7 is characterized in that the described predetermined factor is recorded in the described recording equipment.

10. method according to claim 7 is characterized in that described timer setting is to select from predetermined sampling interval tabulation.

11. method according to claim 7 is characterized in that the step of described analysis histogram data comprises the step of determining the histogram distribution width.

12. method according to claim 7 is characterized in that the step of described analysis histogram data comprises, if described distribution is recognized as narrow distribution, then reduces the step of described data acquisition sample rate.

13. method according to claim 7 is characterized in that the step of described analysis histogram data comprises, is wide distribution if described distribution is recognized as, and then increases the step of described data acquisition sample rate.

14. method according to claim 7, the step that it is characterized in that determining the histogram distribution width comprises the step of counting statistics parameter sets.

15. method according to claim 7 is characterized in that described acoustic environment parameter comprises:

The slope of at least one sound spectrum of-described input signal data;

The modulation of-described input signal data; And

The sound pressure level of-described input signal data.